Abstract
In this paper we propose a new approach to the constructive mathematical representation of nonlinear systems transforming stochastic signals. The approach is based on a combination of a new best approximation technique and a new iterative procedure. For each iteration, the approximation is constructed as a polynomial operator of degree r which minimizes the mean–squared error between a desired output signal and the output signal of the approximating system. We show that this hybrid technique produces a computationally efficient and flexible method for modelling of nonlinear systems. The method has two degrees of freedom, the degree r of the approximating operator and the number of iterations, to decrease the associated error.
Similar content being viewed by others
References
W.A. Porter, “Nonlinear Systems in Hilber Space,” Int. J. Contr, vol. 13, 1971, pp. 593–602.
W.A. Porter, “An Overview of Polynomic System Theory,” Proc. IEEE, vol. 64, 1976, pp. 36–44.
I.W. Sandberg, “Approximately– Finite Memory and Input– Output Maps,” IEEE Trans. on Circuits and Systems. Part 1, Fundamental Theory and Applications, vol. 39, July 1992, pp. 549–556.
I.W. Sandberg, “Criteria for Iniform Approximatiion Using Dynamical Neural Networks,” in Proc. 4th IEEE Mediterranean Symp. Control Automation, Crete, Greece, June 1996, pp. 1–5, Plenary address.
I.W. Sandberg, “Separation Conditions and Approximation of Continuous-Time Approximately Finite Memory Systems,” IEEE Trans. on Circuits and Systems. Part 1, Fundamental Theory and Applications, vol. 46, 1999, pp. 820–829.
I.W. Sandberg, “Time-Delay Polynomial Networks and Quality of Approximation,” IEEE Trans. on Circuits and Systems. Part 1, Fundamental Theory and Applications, vol. 47, 2000, pp. 40–49.
P.G. Howlett, C.E.M. Pearce, and A.P. Torokhti, “On Nonlinear Operator Approximation with Preassigned Accuracy,” J. Comput. Anal. and Applic., vol. 5, 2003, pp. 273–297.
P.G. Howlett and A.P. Torokhti, “A Methodology for the Constructive Approximation of Nonlinear Operators Defined on Noncompact Sets,” Numer. Funct. Anal. and Optimiz., vol. 18, 1997, pp. 343–365.
P.G. Howlett and A.P. Torokhti, “Weak Interpolation and Approximation of Non– Linear Operators on the Space C([0, 1]),” Numer. Funct. Anal. and Optimiz., vol. 19, 1998, pp. 1025–1043.
P.G. Howlett, A.P. Torokhti, and C.E.M. Pearce, “The Modelling and Numerical Simulation of Causal Non–Linear Systems,” Nonlinear Analysis: Theory, Methods & Appl., vol. 47, no. 8, 2001, pp. 5559–5572.
P.G. Howlett, A.P. Torokhti, and C.E.M. Pearce, “A Philosophy for the Modelling of Realistic Non-linear Systems,” Proc. of Amer. Math. Soc., vol. 131, 2003.
A.P. Torokhti, “On Polynomial Synthesis of Non–Linear Systems,” Electronic Modelling. J. of the USSR Acad. of Sci., vol. 11, 1989, pp. 28–4.
A.P. Torokhti, “Modelling of Non-Linear Dynamical Systems,” Electronic Modelling. J. of the USSR Acad. of Sci., vol. 6, 1990, pp. 10–16.
A.P. Torokhti, “On Constructive Restoration of Continuous Mappings in Banach Spaces,” Sov. Math., vol. 1, 1990, pp. 86–89.
A.P. Torokhti, “On Constructive Approximation of Non-Linear Operators,” Sov. Math., vol. 7, 1991, pp. 24–28.
A.P. Torokhti and P.G. Howlett, “On the Constructive Approximation of Non–Linear Operators in theModelling of Dynamical Systems,” J. Austral. Math. Soc. Ser. B, vol. 39, 1997, pp. 1–27.
L. De Lathauwer, B. De Moor, and J. Vanderwalle, “On the Best Rank-1 and Rank-(R1,..., RN) Approximation of Higher-Order Tensors,” SIAM J. on Matrix Anal. and Appl., vol. 21, 2000, 1324–1342.
A.P. Torokhti and P.G. Howlett, “On the Best Quadratic Approximation of Nonlinear Systems,” IEEE Trans. on Circuits and Systems. Part I, Fundamental Theory and Applications, vol. 48, 2001, 595–602.
A. Torokhti and P. Howlett, “Method of Recurrent Best Estimators of Second Degree for Optimal Filtering of Random Signals,” Signal Processing, vol. 83, 2003, pp. 1013–1024.
A.P. Torokhti and P.G. Howlett, “Best Approximation of Operators in the Modelling of Nonlinear Systems,” IEEE Trans. On Circuits & Systems. Part I, Fundamental Theory & Appl., vol. 49, 2002, pp. 1792–1798.
E.I. Lehmann, Testing Statistical Hypotheses, New York: Wiley, 1986.
V.N. Vapnik, Estimation of Dependences Based on Empirical Data, New York: Springer–Verlag, 1982.
M. Jansson and P. Stoica, “Forward-Only and Forward-Backward Sample Covariances-A Comparative Study,” Signal Processing, vol. 7, 1999, pp. 235–245.
P.G. Howlett, C.E.M. Pearce, and A.P. Torokhti, “A Best Linear Estimator for Random Vectors with Values in Hilbert Space,” Maths. Res. (St. Petersburg), vol. 4, 1999, pp. 99–107.
P.G. Howlett, C.E.M. Pearce, and A.P. Torokhti, “Best Estimators of Second Degree for Data Analysis,” Appl. Stoch. Models and Data Anal., Proc. 10th Int. Symp. June 12–15, 2001. Université de Technologie de Compiègne, France, vol. 2, no. 2, 2001, pp. 549–560.
Y. Hua and W.Q. Liu, “Generalized Karhunen–Loève Transform,” IEEE Sig. Proc. Letters., vol. 5, 1998, pp. 141–143.
A.P. Torokhti and P.G. Howlett, “An Optimal Filter of the Second Order,” IEEE Trans. on Signal Proc., vol. 49, 2001, pp. 1044–1048.
A.P. Torokhti and P.G. Howlett, “Optimal Fixed Rank Transform of the Second Degree,” IEEE Trans. On Circuits & Systems. Part II, Analog & Digital Signal Proc., vol. 48, 2001, pp. 309–315.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Torokhti, A., Howlett, P. & Pearce, C. Method of Hybrid Approximations for Modelling of Multidimensional Nonlinear Systems. Multidimensional Systems and Signal Processing 14, 397–410 (2003). https://doi.org/10.1023/A:1023538920581
Issue Date:
DOI: https://doi.org/10.1023/A:1023538920581